Shape memory alloys having one-way or two-way shape memory properties and superelasticity (pseudoelasticity), such as Ni—Ti alloys, Cu—Zn—Al alloys and Fe—Mn—Si alloys, are put into practical use, and most mass-produced are Ni—Ti alloys having excellent properties such as shape memory properties, mechanical strength, etc. However, the Ni—Ti alloys are disadvantageous in poor cold workability, a high material cost, etc. The Cu—Zn—Al alloys have poor corrosion resistance and suffer a high working cost.
As compared with these nonferrous shape memory alloys, iron-based shape memory alloys having a low material cost and good workability are expected to be used for various applications. However, iron-based shape memory alloys developed so far have much poorer superelasticity than that of the nonferrous shape memory alloys, not suitable for applications utilizing superelasticity.
Why conventional iron-based alloys do not have good superelasticity appears to be due to the fact that plastic strain such as dislocation is introduced, and that irreversible martensite (lenticular martensite) which does not have shape memory properties and superelasticity is stress-induced by deformation. To solve these problems, the strengthening of matrix, particularly precipitation strengthening by intermetallic compounds, has been considered effective. From this point of view, an Fe—Ni—Co—Al—C alloy (JP 03-257141 A), an Fe—Ni—Al alloy (JP 2003-268501 A), and an Fe—Ni—Si alloy (JP 2000-17395 A) were proposed. However, even these iron-based shape memory alloys are not necessarily satisfactory in a recoverable strain due to superelasticity, a recovery ratio, superelastically operable temperatures, etc. for practical applications.
“Scripta Materialia” Vol. 46, pp. 471-475 proposes an Fe—Pd alloy containing a large amount of expensive Pd and having a superelasticity. In this alloy, however, the amount of a recoverable strain due to superelasticity is as small as 1% or less.
JP 09-176729 A discloses an Fe—Mn—Si-based alloy utilizing fcc/hcp transformation to exhibit shape memory properties and superelasticity. However, because this Fe—Mn—Si-based alloy exhibits superelasticity only at a higher temperature than room temperature, it cannot be used at room temperature. In addition, because this alloy has poor corrosion resistance and cold workability, needing complicated working and heat treatment, resulting in a high production cost.
U.S. Pat. No. 5,173,131 discloses an iron-based shape memory alloy having a composition comprising 9-13% by weight of Cr, 15-25% by weight of Mn, and 3-6% by weight of Si, the balance being Fe and inevitable impurities, which meets 1.43 (% Si)+1 (% Cr)≦17. In this iron-based shape memory alloy, the difference between a martensitic transformation temperature (Ms) and a reverse transformation temperature (Af) measured by DSC is 110° C. However, this iron-based shape memory alloy is not necessarily satisfactory in a recoverable strain due to superelasticity and a recovery ratio for practical applications.